On 11/12/20 at 13:45 +0100, Michael Biebl wrote: > Am 11.12.20 um 09:26 schrieb Lucas Nussbaum: > > I tried to reproduce this on another system (ARM64, 256 visible cores > > because 2 x ThunderX2, 32 cores/cpu, 4 threads/core) and it also > > segfaults. > > > > I tried to reproduce on yet another system (x86_64, 128 visible cores > > because 4x Intel Xeon Gold 6130, 16 cores/cpu, 2 threads/core), and it > > also segfaults. > > > > My guess would be that the test suite depends on the system it is > > running on (it just generates the output for the local system), and > > there's something it doesn't like about large numbers of cores, but > > there's nothing specific to ppc64el here. > > There is also > $ grep MAXCPUS -R > src/bootchart.h:#define MAXCPUS 16 > src/bootchart.h: double runtime[MAXCPUS]; > src/bootchart.h: double waittime[MAXCPUS]; > src/store.c: if (r < 0 || c > MAXCPUS -1) > src/store.c: /* Oops, we only have room for > MAXCPUS data */
Indeed, it also crashes on an x86_64 system with only 64 visible cores (but not on one with 36 visible cores) Lucas
